What Is a Tech Company?
stratechery.com
stratechery.com
In contrast, there was a company that almost did the same business as we did. But, they were what people would have called a sales company, not a tech company. Decisions were based on sale opportunities, only. No one gave a shit about code quality, devs didn’t stay long and often were frustrated when they applied at our company.
Which one was more successful? Unfortunately they were... we were often slowed down by discussions about code quality, ways of working, ethics and disrespect for revenue driven decisions.
Maybe it was just an exception? But since then I have become skeptical when I hear about “tech” companies. I feel we nerds need to be value driven, too, and not die in beauty.
I think thats the one that hurt you the most.
https://www.kalzumeus.com/2011/10/28/dont-call-yourself-a-pr...
Code quality and developer experience matters because it enables you to quickly use a codebase to solve business problems. But you still have to focus on business problems.
Your developer driven company may have great code quality and employee satisfaction but unless it has a FAANG-like niche all to itself, it will easily be beaten by the revenue driven agile sweatshop next door despite its low quality code and high turnover.
I do think it's easy to go too far the other way, though. You can plan to throw one away. You can write a quick and dirty MVP. You can throw money at the scalability problem if you're lucky enough to have it. Sooner or later, though, you are the player with the established offering. Unless you want to keep rewriting it from scratch every few months, at greater and greater cost as your offering grows, you need decent code at that point to maintain reasonable progress in supporting whatever your next revenue-driven features might be.
That caught my eye as well.
There's such a fine line to walk there where you avoid bad decisions "because customer X wants it" and recognition that giving the customer what they want is what you are there to do / what sells.
I just had a discussion that was "We're not going to do this how they want as that is too wonky / would send the code in all sorts of strange directions ... but we're still going to give them what they want in spirit."
If we had just said no, or blown off the sales guy's POV, probabbly lose a lot of traction with that customer.
1. The business and product
2. The engineering quality
You can have a tech company (business level as described in the article) and produce ugly code (ex: facebook?).
You can sell groceries on the internet and have great code.
Revenue and business impact are ALL that matter.
Good, clean, scaleable, and testable code only exist to rapidly support the above. The moment you choose tech value over buisness you are a risk not asset to a company.
I sware sometimes I see so many delusional software "engineers" who care more about building technically excellent systems that are either delivered late or dont solve a/the real problem... it disgusts me.
> Revenue and business impact are ALL that matter.
Everywhere else around the world software development is just a SCRUM powered meet grinder where if you don't sprint fast and cheap enough your competitors will eat your lunch.
Unfortunately, shiny engineering won't solve drastic product market fit iterations, and consumes the limited time & $$$ for achieving it. That's a lot of otherwise good habits and instincts to unlearn for engineers who never really had to go from idea to paying end users and instead needed to ride an existing monopoly and not break it: FANG, bigco, bank, consultants, etc. In a disturbing sense, a small # of new grads can fit in better on a startup team than a loud FANG cohort b/c most know how to do a bunch of small projects without being distracted by post-growth table stakes. (And you avoid the former-consultant "don't care about the end-customer, just the boss/politics" problem b/c of passion for the startup's mission vs. the tech.)
I've come to appreciate either:
Team 1: Staying tiny with people who demonstrably get that or are otherwise mission-focused (vs. tech-focused) until you do hit growth,
Team 2: Or massively over-fundraising and over-hiring (with high risk of fast death) simply to deal with sales/marketing/engineering that is only at say 10% efficiency of Team 1
>Unfortunately, shiny engineering won't solve drastic product market fit iterations, and consumes the limited time & $$$ for achieving it. That's a lot of otherwise good habits and instincts to unlearn for engineers who never really had to go from idea to paying
Having worked in startups and the corporate world, I would apply this also to the role of product owner, who is often responsible for setting up those experiments to learn and also filling up the tech pipeline for the next sprint. I've seen product owners having a hard time to grasp if the technical need for things pitched by engineering is real or just premature optimization. You either need a more technical product owner or a tech lead that puts it's business hat on quite often.
> I've come to appreciate either...
I think there's room for balanced consideration. Sometimes, yeah, revenue and time-to-market is everything. Sometimes there's a need for taking a risk and creating something new.
The new business is the risk-taking endeavor. The job of the team is to derisk, taking the blue-sky concept to something the market will consume. That's so hard and requires so much listening/iterating/etc. that it makes most engineering cycles not spent on meeting customers in the middle (or helping finding alternate customers/offerings) is eating the seed corn. Most startups should assume "default dead" until the $5M/yr point.
There is an interesting question of how to do deep tech. ("10X cheaper spaceflight", not "Tensorflow for AIOps"). That's more capital intensive, and in turn, different kinds of business critical milestones. "We need to do X, Y, Z so we can get $5M gov contract Y and the next round." That deal does _not_ hinge on whether you've upgraded from React 3 to 4 and have a much more approachable graphql orm.
If some MBA with PMP certs decides to actively ignore the business impact of your bad code, it means he is stupid and incompetent. But that does not mean that code quality is an end on its own.
You might want to look up b-corporations if you haven't. They take into account employees, communities, and the environment as parts of their charters.
I'd heard about them for a while but didn't realize the difference they made until speaking to Lorna Davis, who was instrumental in the largest conversion to a b-corp yet.
I don't claim it's the best introductions to them, but the conversation was on my podcast, so you can hear her personal inside account: http://joshuaspodek.com/guests/lorna-davis.
Isnt that what google did to yahoo? Apple to Nokia? And now kubernetes does to AWS?
For a monster like Amazon it particularly doesn't make sense to call it a technology company or an ecommerce company (or a media company or an ad company). It's a company - really a conglomerate - and it does/sells x y z. AWS is as much a critial part of what Amazon is today as its retail business (on a value basis, AWS is drastically more valuable, and will be even more so in the future). One day they may be split apart, however that is likely a decade into the future or more.
That's a management failure not software engineers doing the wrong thing. Software engineers should care more about building technically excellent systems just like safety auditors should care most about safety. It's management's job to prioritize the wants of the business and the wants of the engineers and decide when they align and when they don't.
Er, they also exist to support the quality of life of the engineer. Suppose your company asks you to spend 6 months writing PHP in an untestable codebase. And you know that you need to have testable code in order to make progress in a codebase. It is better for you to stay off that project, even if it theoretically would make massive revenue gains for the company.
Why?
Because if the company's revenue doesn't matter to you when annual review time rolls around and you get fired because you didn't accomplish much in the past 6 months. So yes, you should care about business goals, but only insofar as they support the stability of your family life.
1. Notice how the author mentions the term "tech industry"? However, when most people talk about "tech companies" they refer to it as an industry. It's not.
2. People get lost in the semantics and just think companies are a binary "tech co" or "not a tech co". I disagree. Tech is actually an operating model which can help propel a company into high gross margins. When we analyze businesses we usually put them on a spectrum of "IT Support Business" -> "Tech Enabled Business" -> "Tech Lead Business" (with the latter synonymous with that people call a "tech company").
3. A "Tech company" and a "software company" is nothing more than an investors way to imply underlying business fundamentals, which usually translate into high gross margins. It can often be used to cloud judgement about the actual underlying fundamentals.
* Software creates ecosystems
* Software has zero marginal costs
* Software improves over time
* Software offers infinite leverage
* Software enables zero transaction costs
The article says this about Uber: "Uber is able to serve the entire world, giving it maximum leverage". Followed by this about WeWork: "WeWork is limited by the number of locations it builds out".
Uber didn't start with the ability to serve the entire world, and that ability was earned not through software, but through lots and lots of money and aggressive regulatory arbitrage (i.e. breaking laws faster than regulators could react to). What's to say that WeWork can't spend money to grow into a worldwide service?
I see this article as a list of interesting market features, but I don't buy that you have to have all of them to be a tech company, or that only companies with all of them will succeed.
"Infinite" leverage is also a misnomer: Ford and GE enjoyed terrific leverage when they optimized giant-scale industrial processes (the assembly line and six sigma, respectively). Politicians and lawyers enjoy much greater leverage than software engineers, whose code requires much more "maintenance" (in practice) than laws and legal contracts. Economically speaking, Wall Street traders enjoy fabulous leverage, though arguably they're getting replaced by algorithms.
Fifty years ago, IBM was the computer company. There were other tech companies: Xerox, Tektronix, Texas Instruments, and Western Electric to name a few. Today "tech" largely means computing, especially applied to marketing and advertising automation, but back then it meant something quite different to be working "in technology".
For example, what about companies like DuPont. Surely they are a technology company. Are they also a tech company? The author provides no framework to address the question which, in my mind, means they were largely unsuccessful in their attempt.
For instance, uber is a transportation company, not a tech company. They may have been considered tech when they were developing self-driving cars but I think that was abandoned for now?
Anyway, the point is that being in the internet and having an app is par for the course of any industry today. You're not on the brave frontier of tech anymore, you're just using existing tech to implement your business model.
If you're making NEW tech, you're a tech company.
Technology is "the application of scientific knowledge for practical purposes, especially in industry."
People tend to confuse tech and tools. The lever is a tool. Using a lever to create leverage, is technology. (You can argue that the lever is technology too, in that its applied knowledge of smelting, but that misses the point.)
Uber is the application of engineering talent to automate taxi dispatch. Netflix is the application of engineering talent to distribute video. But that doesnt make them tech companies, because they dont SELL or LICENSE that tech to others. Their product itself isnt tech for others to use. They have in housed software, and they use their software development prowess as a competitive advantage. "Tech companies" should be reserved for the Microsoft, Google, Apple, Salesforce, SAPs of the world that SELL/LICENSE software that acts as a Business Middleware and Development Platforms. But more importantly, they need to be MORE than just a company that takes business requirements and digitizes them (UltiPro for example, who applies existing Microsoft/Google tech/tools, but isnt creating their own database tech or frameworks. You can be a software assembly company, that solves business problems, without being a tech company,) they need to be inventive, creating new applied knowledge. (In that case, Uber actually does tangentially count, because they are a software middleware to connect drivers and riders, and they are engineering new tech not just reassembling off the shelf components.)
For nearly everything else, we can just call a spade a spade and call a company what it actually is. A bank. A car company. Taxi dispatch. Video hosting and production. Semiconductors. Fabrication. Manufacturing.
Lumping every company that has good engineering talent into a pile may be good for investors to help categorize investments, but for everyone else it just turns conversation into nonsense.
Oftentimes the label tech company is used as shorthand for Internet company — a company that leverages the near-zero distribution cost of the Internet to deliver a product (e.g. Google Search, Facebook) or a service (e.g urban mobility via Uber, home delivery via Amazon).
In this sense, it would not be correct to say Uber or Netflix are not tech companies.
Microsoft, Google, Apple, Amazon build databases, programming frameworks, hardware, operating systems. "Tech platform vendor?"
I do think we should limit its use to something of a legal term, as I tried to explain elsewhere in the thread: https://news.ycombinator.com/item?id=20879152
Rather, they provide a post-paid software platform for individuals to provide transportation services to customers directly.
Delta is a transportation company because Delta is cutting the checks for the purchases and maintenance of the airplanes.
Sure, it's a spectrum, but Delta isn't a tech company. Skyscanner and Booking.com are tech companies that fulfill an Uber-like service for Delta.
Uber isn't saying, "you must drive from 9am to 5pm. you must use our vehicle and company expense account."
Mechanics get hit with these types of laws as well. Mechanics are often required to bring their own tools to work, but since they often work at the business' garage during the business' hours (not at their home garage at 2am for example), they are W2.
You are exactly right that the service rendered is exactly the same ("I wanna uber there" == "I want a taxi company to drive me there").
But from an expenditure standpoint, a Taxi Company is not spending money on cloud services and programmer salaries.
My view is that you're in the business that customers are paying you for. You may have been founded by tech-heavy leadership and have a heavy reliance on software, but that doesn't define what people actually pay you for. No more than a company founded by bankers is a bank.
Exactly. In fact, the most useful explanation I’ve found is to consider the label “tech company” as mostly a legal term, similar to how a bank is a legal label for a specific type of company (with the small difference that banks must be licensed by the government to conduct the activity of banking).
To intuitively understand the term tech company, consider the level just below it: the tech-enabled company. The distinction between a tech company and a tech-enabled company is exactly the same as the (legal) distinction between ownership and possession.
Tech companies generally build (i.e own) the primary means of production (i.e. tools, methods, raw materials etc used to conduct business). Exclusive ownership of a valuable means of production is generally a source of competitive advantage in most industries, until a new entrant with better economics emerges to disrupt the status quo.
Tech-enabled companies on the other hand, generally buy or borrow (i.e. possess) the primary means of production.
IOW, all companies can be placed on the tech-company spectrum depending on whether they build, buy or borrow the primary technology necessary to conduct business.
An example of a tech company: Boeing; an example of a tech-enabled company: United Airlines. UA cannot design or manufacture its own aircraft since it does not own the underlying tech (IPR, manufacturing methods etc) behind the aircraft it operates, instead it merely purchases from tech vendors like Boeing (and Airbus).
If UA decides that battery tech will be viable in the near future and wants to switch its entire fleet to all-electric aircrafts for instance, they must seek the legal permission of their tech vendors to modify their existing designs, since retrofitting will require recertification to ensure the aircrafts can still be operated safely. If they go ahead with such modifications without first seeking permission, they risk being sued by the owner of the underlying technology.
OTOH, if Boeing were to make the same realization regarding battery-powered flight, of course they would not need to seek any third party’s permission to make the switch.
Let’s say UA or indeed any airline wants to take its destiny into its own hands, what prevents them? Surely they could kick start their R&D efforts buying an electric aircraft maker like Pipistrel [3]? UA used to be a vertically integrated company known as UATC [0] but the USG found such an arrangement too powerful and anti-competitive, which led to the split into current day Boeing, UTC [1] and of course UA [2].
0: https://en.wikipedia.org/wiki/United_Aircraft_and_Transport_...
1: https://en.wikipedia.org/wiki/United_Technologies
In that context a "tech company" specifically refers to a company that is able to take advantage of new business models or ways of operation enabled by tech that allows them to scale and make large returns on investment. Thats it. Thats why they are attractive to investors, which in turn is why every company tries to convince people that they are also "tech".
> [Hobbes'] Leviathan was not concerned with idle speculation about new political principles or the old search for reason as it governs the community of men; it was strictly a "reckoning of the consequences" that follow from the rise of a new class in society whose existence is essentially tied up with property as a dynamic, new property-producing device. The so-called accumulation of capital which gave birth to the bourgeoisie changed the very conception of property and wealth: they were no longer considered to be the results of accumulation and acquisition but their beginnings; wealth became a never-ending process of getting wealthier. The classification of the bourgeoisie as an owning class is only superficially correct, for a characteristic of this class has been that everybody could belong to it who conceived of life as a process of perpetually becoming wealthier, and considered money as something sacrosanct which under no circumstances should be a mere commodity for consumption.
-- Hannah Arendt, "Origins of Totalitarianism"
What is missing however in the picture of "someone building a site" is the other points mentioned in the essay: network effects, building a platform and forming an ecosystem around you. Everyone can do it, but not everyone succeeds. This is why there is such a big rush within tech startups to be the first in a space.
I call my startup a software company because that is what it is. Tech is abbreviation from "high tech", which was the hardware and biotech equivalent.
It's a bit more nebulous with e.g. Uber though; they don't really sell software, they sell a service and act as a middle man via software instead.
And other car companies are not? The modern internal combustion engine is a technological marvel, and getting better every year.
Tesla has a bigger market cap than Ford, isn’t profitable and is always one deal alway from insolvency. The value is it’s tech and potential.
Ditto with cars. Telsa is not a tech company. It's a car manufacturer.
Intel is a chip company.
Microsoft is a software company... except there's Microsoft hardware out there, isn't there?
Do you think there might be a use for an over-arching term to talk about all of those companies at the same time, while leaving out, say, Unilever and Coca-Cola?
Numbers 2-4 are clearly tech companies, while Tesla, Coca-Cola, and Unilever clearly are not, no matter how much technology they use in delivering their cars, soda, and household goods.
I think you think 'technology' means 'computer'.
Detergents are also a technology.
Nearly everything sold involves technology. Tech companies sell the technology more directly, while non-tech companies sell the results of technological processes. Generally speaking, yes, this means that tech companies are mostly computer companies. Even companies that provide SAAS could theoretically also sell that software directly (and often do, to large enough enterprise customers), so they're still software companies.
If it weren't for companies muddying the waters while trying to raise large quantities of money, I don't think the distinction would be in any doubt whatsoever.
This is also true for a lot of software companies (boxed software, games, paid upgrades etc.) Not to mention that a lot of companies selling hardware should also qualify as tech.
My big problem is that the author is concluding that tech = software = subscription-based SaaS.
Microcode updates do not add new features, they exclusively fix bugs.
(sorry, couldn't resist)
Software installed locally that doesn't run "in the cloud" would qualify as not tech according to that definition.
For the most part Windows 98 just got bug fixes, in the same way a car might get part recalls, until Win 2000/ME came out with new features.
Software installed locally may still receive updates.
--
[0] - and replacement is fair; software also improves by replacement. Your old executable gets deleted and replaced by a new executable.
>Car
>Mattress
Non tech? What?
What?
... what?
Excluding manufactured goods from tech just kind of broke my brain for a minute there. I guess "tech" and "technology" are different words now and I just have to accept that.
I have a feeling that most people would call Philips Hue a Tech product, and if that was Philips only product we would call them a Tech company.
For what it's worth I don't think that's good and I prefer to call Slack a software company, Uber a taxi company and WeWork a real estate company.
Language evolves, I guess. I dislike this change, since it excludes all the radar systems, jet fighters, lasers and other high technology systems that I've worked on, but there's nothing I can do about it. Oh well.
You get packaged software, you have to buy new version to get new features.
You have SaaS solution, you have to pay subscription fee, so basically for monthly payments you pay for that months version.
Lets make a hypothetical counterexample according to their guidelines:
- Software creates ecosystems.
- Software has zero marginal costs.
- Software improves over time.
- Software offers infinite leverage.
- Software enables zero transaction costs.
Imagine that out of nowhere, we get a company making a JARVIS-like AI assistants like in iron man, complete with google glass-like HUD. Suppose, due to design/manufacturing restrictions or something, it can't update over time. You have to buy a new pair of the glasses to get the improvements. There are no third party apps, just what comes shipped with the product. If it was as useful as envisioned all over science fiction, it'll take the world by storm anyways.
Now imagine it's just catching on with consumers and is about to IPO. Is this AI driven high-tech company a tech company? According to OP, this company may as well be selling sunglasses:
- No apps or ecosystems by construction.
- Every unit is a physical object, marginal costs are about like expensive sunglasses.
- It doesn't improve over time by construction.
- No more leverage than selling normal sunglasses.
- No lower transaction costs than a company selling sunglasses.
Yet clearly this is a tech company and will be evaluated accordingly for its IPO.
Do you think your example fits into the author's definition of "software-differentiated hardware"?
I read it as the author saying "oh by the way these other companies that don't fit are their own (massive) category, not counterexamples to the criteria I've laid out."
I think the answer is partly that you're right and that technology in the more general broad sense has particular tendency to keep improving in value; partly that the OP is right and your example has some unstated implications (these non-upgradeable devices might well be prone to competition and hence generate surprisingly low shareholder returns); and partly that there are some other factors beyond OP's, to do with how the goods are produced.
1. Software is the core product and/or not considered a cost center.
2. Technical decisions are made by technical people.
What these both come down to is that a tech company is one where software engineers are treated as full professionals rather than simply as labor.
(Note that I prefer the old meaning of the word "tech" when it was short for "technology". But it's not up to me.)
The article makes some great points, but I think the use of software and technology for strategic advantage is different from the fundamental "purpose" of the company.
For instance -- Intel is a clearly a tech company, but specifically they are more of a manufacturing company than a software platform provider.
I have worked for certain communication mega-corp, while the business is (somewhat) technical, the dev have almost no say at all. That is where I would draw the line and call that a 'traditional' company. I've since found job at a real tech company.
Completely off-topic, I read somewhere recently that EBITDA could also stand for "Earnings Before I Tricked the Dumb Auditor".
While we're here it is worth remembering that unlike most of today's "rising stars", most of the established tech giants were profitable at IPO time (Google, Facebook, Apple, Microsoft).
Things like WeWork's "community-adjusted EBITDA" read like punchlines, though.
(1) Marketed to investors as a tech company
(2) Valued by investors based on the premise that they are or will become oligopolists in their market
The simplification that doesn't do justice to the article is: Tech companies are those that are more efficient than their competitors.
That's a solid argument for abandoning the term altogether, in my view. If the term could apply to such a gigantic amount of different things, what is the worth of making that distinction?